BE - determine approach to feature flag filings

Determine the general approach of how the BE will use feature flags such that we can turn a given filing on/off based off of specific targeting criteria moving forward.

- [ ] Feature flags targeting criteria
- [ ] environment (dev/test/prod)
- [ ] user types (general user & staff)
- [ ] filing type & filing sub-type
- [ ] legal type for a given filing type/filing sub-type
- [ ] login source (so we can turn on filings in the UI and generally block/allow 3rd party API access)
- [ ] how will the FF be defined in LD? do we need to use JSON?
- [ ] how will the FF integrate with allowable actions?
- [ ] Does the approach work in a way where we can use the same pattern for all current/future filings(including ones with sub-types)?

No files or tests are named. Start by locating the backend's existing feature-flag and allowable-actions integration, then compare it with the listed targeting criteria and LaunchDarkly questions. Done means an agreed, reusable approach covering environments, users, filing and legal types, login source, flag definition, and integration.

Written by the indexing model from the issue text.